Contact
Prof. Dr.-Ing. Jens Kober
University of Stuttgart
Institute for Artificial Intelligence
Learning and Interactive Robots
Universitätsstraße 32
70569 Stuttgart, Germany
🖂 jens.kober (at) ki.uni-stuttgart.de
🕿 +49 711 685 88102
Office: 315a
Fraunhofer-Institut für Produktionstechnik und Automatisierung IPA
Nobelstraße 12
70569 Stuttgart, Germany
🖂 jens.kober (at) ipa.fraunhofer.de
https://www.linkedin.com/in/jens-kober/
Research interests
Robotics, Motor Skill Learning, Machine Learning, Reinforcement Learning, Imitation Learning, Deep Learning, Interactive Learning
News
- 3 accepted conference papers:
. Set-Supervised Diffusion Policy: Learning Action-Chunking Diffusion through Corrections. In Robotics: Science and Systems (R:SS), 2026. Accepted. [bibtex] [pdf] [webpage] [code] [video]
. Evaluating Randomized Smoothing as a Defense Against Adversarial Attacks in Trajectory Prediction. In IEEE 29th International Conference on Intelligent Transportation Systems (ITSC), 2026. Accepted. [bibtex] [pdf] [code]
. Increasing the blade tip-to-tower clearance using individual pitch control. In Torque, 2026. Accepted.
- 2 new preprints:
. Resolving Space-Sharing Conflicts in Road User Interactions Through Uncertainty Reduction: An Active Inference-Based Computational Model. arXiv:2604.19838 [cs.AI], 2026. [bibtex] [pdf] [doi]
. Search Inspired Exploration in Reinforcement Learning. arXiv:2602.00460 [cs.LG], 2026. [bibtex] [pdf] [doi]
- Jens Kober started a new position as full professor at University of Stuttgart with a joint appointment at Fraunhofer IPA.
- 1 accepted journal paper:
. Accurate and Affordable Cobot Calibration Without External Measurement Devices. Communications Engineering, 2026. in press. [bibtex] [pdf] [doi] [code] [video]
- 1 new preprint:
. TacLight: A Six-Axis Soft Optical Tactile Sensor for Physical and Virtual Interaction. PREPRINT (Version 1) available at Research Square, 2026. [bibtex] [pdf] [doi]
- 2 accepted conference papers:
. LLM-Guided Task- and Affordance-Level Exploration in Reinforcement Learning. In IEEE International Conference on Robotics and Automation (ICRA), 2026. Accepted. [bibtex] [pdf] [webpage] [code] [video]
. Studying the Effect of Explicit Interaction Representations on Learning Scene-Level Distributions of Human Trajectories. In IEEE Intelligent Vehicles Symposium (IV), 2026. Accepted. [bibtex] [pdf] [code] [data]
- 1 accepted journal paper:
. Sequentially Teaching Sequential Tasks (ST)2: Teaching Robots Long-Horizon Manipulation Skills. IEEE Robotics & Automation Magazine, 33(1):130–141, 2026. [bibtex] [pdf] [doi]
- 2 new preprints:
. Controllable Generative Trajectory Prediction via Weak Preference Alignment. arXiv:2510.10731 [cs.RO], 2025. [bibtex] [pdf] [doi]
. STEP: Structured Training and Evaluation Platform for benchmarking trajectory prediction models. arXiv:2509.14801 [cs.LG], 2025. [bibtex] [pdf] [doi] [code]
- 4 accepted journal papers:
. Human-Interactive Robot Learning: Definition, Challenges, and Recommendations. ACM Transactions on Human-Robot Interaction, 15(2):1–31, 2026. Article 46. [bibtex] [file] [doi]
. Scalable Task Planning via Large Language Models and Structured World Representations. Advanced Robotics Research, :e202500002, 2025. [bibtex] [pdf] [doi] [video]
. ILeSiA: Interactive Learning of Situational Awareness from Camera Input. IEEE Robotics and Automation Letters, 10(10):10490–10497, 2025. [bibtex] [pdf] [webpage] [doi] [code] [video]
. ASkDAgger: Active Skill-level Data Aggregation for Interactive Imitation Learning. Transactions on Machine Learning Research, 2025. [bibtex] [pdf] [url] [webpage] [code] [video]